With the budget well in the rear-view mirror, Jim Chalmers has increasingly become chief putter of lipstick on the Australian economic pig. The Treasurer and his small army of media and political advisers – and perhaps the bulk of his 1600-odd staff at the Treasury – must dread the drip feed of conventional economic statistics. And so it was last week when the latest national accounts showed gross domestic product grew 0.4 per cent in the three months to the end of June, to finish the full financial year up 2.1 per cent.
According to the Treasurer, this demonstrated the economy’s robustness and resilience, as every set of figures always has for every treasurer. But they were a dismal set of results. The quarterly “growth” was propped up by government spending while private investment didn’t increase at all. Per-person output barely rose and it is still lower than in June 2022, having shrunk in 10 of the past 16 quarters.
But the bigger problem is the ongoing obsessive focus on GDP. Twelve years ago I wrote GDP was “a harmless artefact of a bygone era and at worst a ruse under pinning the fallacy that bureaucrats can measure and politicians can control economic activity”. People cite GDP, which is increasingly made up of government spending, without even understanding it; and its shortcomings have become even more glaring as a measure of anything useful.
GDP has evolved from one useful measure of economic production decades ago into the universal yardstick against which almost everything else is judged: debt, greenhouse emissions, productivity and even share market and housing values.
I’m not the only sceptic. Barely a day goes by when some politician or analyst decries the relentless increase in debt to GDP somewhere in the world. “We’re becoming Argentina” is the warning.
It’s a funny claim given Argentina defaulted in 2001 when its debt to GDP was about 60 per cent. Meanwhile, Japan’s ratio now exceeds 250 per cent and America’s 100per cent. Britain’s debt-to-GDP ratio has roughly doubled over the past 40 years to around 95 per cent.
This all sounds rather scary but using government interest repayments as a share of GDP or national financial wealth paints a dramatically less alarming picture, according to economists Jonathan Berk and Jules van Binsbergen in their recent paper, Why Care About Debt to GDP?
“While the debt-to-GDP ratio has recently surpassed levels of indebtedness not seen since World War II, the interest-to-GDP ratio reached comparable levels during the1985-1995 period. Most strikingly, the debt-to-equity ratio appears to show no clear trend and, if anything, has recently fallen to below-average levels,” they write.
Different measures produce very different narratives. In another recent study, When GDP Misleads, released in June, Stanford economist Charles Jones and Philip Trammell argue against using real GDP per person as a proxy for living standards.
GDP per capita has been used to show how Europeans’ living standards have shrunk below even those of Alabama, the poorest US state. Yet if your chosen measure implies life in impoverished, fentanyl-riddled Mobile is superior to that in Barcelona, it tells us more about the statistical poverty of GDP per capita than living standards.
Using a broader method of analysis related to the value of a statistical life, based on individuals’ willingness to pay for a small reduction in mortality risk, the two economists find that lifetime welfare in the US has increased more than fivefold since 1940. Conventional GDP-based measures, by contrast, imply living standards have around doubled since then.
Their assessment suggests conventional GDP-style measures may have massively understated improvements in living standards.
Remember, more purchases of goods and services is “good for GDP”. Yet the modern smartphone has replaced the camera, video recorder, encyclopaedia, street directory, calculator, music player, alarm clock, torch, diary, newspaper and countless other devices that previous generations purchased separately. Free digital services that millions of people now use every day generate no economic output at all. “A natural implication of this structural transformation (toward such services) is that measured GDP would likely be substantially higher if we had never made the structural transformation.”
This should be the death knell of GDP as a measure of living standards. And that’s before we even consider the rapid march of artificial intelligence through the economy. Jones, one of the world’s top economic historians, argues AI is likely to become the most transformative technology of the modern era, but it won’t appear in GDP per capita.
Technological capability can advance at breathtaking speed while measured productivity and GDP rise only gradually. Artificial intelligence seems likely to magnify that phenomenon.
Imagine if every household has access to a highly capable AI tutor, accountant, software engineer, travel agent, translator and preliminary medical adviser at little or no cost.
Perhaps GDP has gradually become something it was never intended to be. It has evolved from one useful measure of largely private sector economic production into economics’ universal scoreboard, a statistic expected simultaneously to measure output, prosperity, living standards and even social progress. There are far better measures for all those other things.
